Original Research ajog.

org

GYNECOLOGY
Uncovering changes in proteomic signature of rat pelvic
floor muscles in pregnancy
Lindsey A. Burnett, PhD, MD; Francesca Sesillo Boscolo, PhD; Louise C. Laurent, MD, PhD; Michelle Wong, BS;
Marianna Alperin, MD, MS

BACKGROUND: Structural and functional changes of the rat pelvic
floor muscles during pregnancy, specifically, sarcomerogenesis, increase
in extracellular matrix content, and higher passive tension at larger strains
protect the integral muscle components against birth injury. The mech-
anisms underlying these antepartum alterations are unknown. Quantitative
proteomics is an unbiased method of identifying protein expression
changes in differentially conditioned samples. Therefore, proteomics
analysis provides an opportunity to identify molecular mechanisms un-
derlying antepartum muscle plasticity.
OBJECTIVE: To elucidate putative mechanisms accountable for
pregnancy-induced adaptations of the pelvic floor muscles, and to identify
other novel antepartum alterations of the pelvic floor muscles.
MATERIALS AND METHODS: Pelvic floor muscles, comprised of
coccygeus, iliocaudalis, and pubocaudalis, and nonpelvic limb muscle,
tibialis anterior, were harvested from 3-month-old nonpregnant and late-
pregnant Sprague-Dawley rats. After tissue homogenization, trypsin-
digested peptides were analyzed by ultra-high-performance liquid chro-
matography coupled with tandem mass spectroscopy using nano-spray
ionization. Peptide identification and label free relative quantification
analysis were carried out using Peaks Studio 8.5 software (Bioinformatics
Solutions Inc., Waterloo, ON, Canada). Proteomics data were visualized
using the Qlucore Omics Explorer (New York, NY). Differentially expressed
peptides were identified using the multi-group differential expression
function, with q-value cutoff set at <0.05. Proteomic signatures of the
pelvic floor muscles were compared to nonpelvic limb muscle and
between nonpregnant and pregnant states.
RESULTS: Unsupervised clustering of the data showed clear
separation between samples from nonpregnant and pregnant ani-
mals along principal component 1 and between pelvic and nonpelvic
muscles along principal component 2. Four major gene clusters
were identified segregating proteomic signatures of muscles exam-
ined in nonpregnant vs pregnant states: (1) proteins increased in the
pelvic floor muscles only; (2) proteins increased in the pelvic floor
muscles and tibialis anterior; (3) proteins decreased in the pelvic
floor muscles and tibialis anterior; and (4) proteins decreased in the
pelvic floor muscles alone. Cluster 1 included proteins involved in
cell cycle progression and differentiation. Cluster 2 contained pro-
teins that participate in mitochondrial metabolism. Cluster 3 included
proteins involved in transcription, signal transduction, and phos-
phorylation. Cluster 4 comprised proteins involved in calcium-
mediated regulation of muscle contraction via the troponin tropo-
myosin complex.
CONCLUSION: Pelvic floor muscles gain a distinct proteomic
signature in pregnancy, which provides a mechanistic foundation for
the antepartum physiological alterations acquired by these muscles.
Variability in genes encoding these proteins may alter plasticity of the
pelvic floor muscles and therefore the extent of the protective
pregnancy-induced adaptations. Furthermore, pelvic floor muscles’
proteome is divergent from that of the nonpelvic skeletal muscles.
KEY WORDS: pelvic floor muscles, pregnancy adaptations,
proteomics, rat

P elvic floor disorders, including

pelvic organ prolapse and urinary
and fecal incontinence, are exceedingly
common conditions, with a prevalence
of 23.7% among community-dwelling
women in the United States.1 These
chronic disorders have a tremendous
negative impact on quality of life. Pelvic
floor muscle (PFM) dysfunction is a
major risk factor for the development of
pelvic floor disorders, especially pelvic
Cite this title as: Burnett LA, Sesillo Boscolo F, Laurent
LC, et al. Uncovering changes in proteomic signature of
rat pelvic floor muscles in pregnancy. Am J Obstet
Gynecol 2019;XX:x.ex-x.ex.
0002-9378/$36.00
ª 2019 Elsevier Inc. All rights reserved.
https://doi.org/10.1016/j.ajog.2019.04.025
organ prolapse. Although multiple pre-
disposing and promoting factors have
been identified, the single most signifi-
cant event accountable for PFM
dysfunction is vaginal delivery. During
vaginal birth, PFMs are hypothesized to
elongate up to 300% of resting length.2,3
In the limb skeletal muscles, such strains
consistently result in muscle injury,
suggesting that all vaginally parous
women should sustain PFM injury.4
Surprisingly, only w30% of vaginally
parous women demonstrate radiologi-
cally visible PFM injury.5 PFMs likely
undergo adaptations that change muscle
physiological limits to facilitate fetal de-
livery, while protecting against maternal
injury. Given ethical constraints associ-
ated with directly probing these muscles
in living women, we rely on direct tissue-
level studies of PFMs in animal models
to help us uncover the reasons for the
above. The rat model is used in this
study, as it has been previously demon-
strated to be an excellent model for tissue
changes in pregnancy,6e8 and the rat
pelvic floor muscle anatomy and archi-
tecture are similar to those of humans.9,10
Our previous investigations have also
demonstrated protective pregnancy-
induced adaptations of the rat
PFMs that favorably alter muscle
response to strains associated with de-
livery.11 These adaptations, specifically,
sarcomerogenesis, increase in intramus-
cular extracellular matrix content, and
higher passive tension at larger strains,
appear to be critical to the PFMs’ ability
to withstand excessive parturition-
related strains by attenuating sarcomere

MONTH 2019 American Journal of Obstetrics & Gynecology 1.e1

Original Research GYNECOLOGY ajog.org

phosphate-buffered saline solution and

AJOG at a Glance incubated overnight at room tempera-
Why was this study conducted? ture. Samples were then centrifuged, and
This study elucidates putative mechanisms underlying protective pregnancy- pelleted solids were resuspended in 1%
induced adaptations in rat pelvic floor muscles and identifies novel antepartum sodium dodecyl sulfate in phosphate-
proteomic alterations of pelvic muscles. buffered saline solution and then dis-
rupted by brief homogenization. Next,
Key findings tryptic digest and peptide isolation were
 Pregnancy increases pelvic floor muscles’ expression of proteins that stimulate performed by filter-aided sample prep-
cell proliferation and decrease apoptosis. aration, as previously described.18 In
 Expression of fast skeletal muscle protein isoforms decreases in pelvic floor brief, samples in 4% sodium dodecyl
muscles during pregnancy, suggesting fast to slow fiber type transition. sulfate and 10 mM dithithreotide were
 Muscle metabolism in pregnancy favors aerobic respiration with increased applied to spin filters, incubated at room
expression of oxidative enzymes and mitochondrial proteins, consistent with temperature for 10 minutes, and boiled
the transition in fiber phenotype. at 100 C for 5 minutes. Spin columns
were washed with 0.1 M Tris 8M urea.
What does this add to what is known? Samples were carboxymethylated with
We identified candidate modulators of protective pregnancy-induced adapta- 0.5 mg/mL of iodoacetamide for 20 mi-
tions of rat pelvic floor muscles, specifically pathways involved in muscle resident nutes at 37 C in the dark and then
stem cell component and muscle metabolism. Understanding antepartum washed with urea solution. Tryptic digest
changes in pelvic muscles is essential to harness their endogenous potential to was performed on spin filter with 0.03
prevent or mitigate maternal birth injury. mg/mL trypsin in 50 mM ammonium
bicarbonate overnight at 37 C. Spin fil-
hyperelongation and thus preventing to identify other novel antepartum ters were washed with 50 mM ammo-
muscle injury.11e13 Consequently, a alterations of PFMs. nium bicarbonate, and peptides were
deeper understanding of the cellular and eluted with 0.5 M sodium chloride.
molecular mechanisms underlying these Materials and Methods Elutant was dried with speed vac and
structural and functional adaptations is Muscle procurement resuspended in 0.5% trifluoracetic acid
needed to be able to harness the endog- The University of California San Diego and 5% acetonitrile. Samples were zip
enous plasticity of PFMs. Building on Institutional Animal Care and Use tipped (Millipore, Burlington, MA) per
our previous studies, we sought to obtain Committee approved all study proced- the manufacturer’s instructions.
a global understanding of the rat PFMs’ ures. Nulligravid nonpregnant (n ¼ 3)
response to pregnancy, by using an un- and primigravid late pregnant (n ¼ 3) Mass spectroscopy
biased quantitative proteomic approach. 3-month-old Sprague-Dawley rats were Trypsin-digested peptides were analyzed
Quantitative proteomics offers an op- obtained from Envigo (Indianapolis, by ultra-high-pressure liquid chroma-
portunity to simultaneously identify and IN). Nonpregnant animals were in tography coupled with tandem mass
quantify thousands of proteins within a similar parts of the estrous cycle as spectroscopy using nano-spray ioniza-
tissue. With quantitative mass spec- determined by vaginal smear. Pregnant tion. The nanospray ionization experi-
trometry analysis, identification of rela- rats were in day 20e21 of gestation. ments were performed using a TripleTof
tively small changes in protein expression Animals were sacrificed, and pelvic 5600 hybrid mass spectrometer
between different conditions can be floor muscles, including coccygeus (C) (ABSCIEX) interfaced with nano-scale
accomplished. As a result, small but and the individual components of the reversed-phase ultra-high-pressure
biologically significant protein alterations levator ani, pubocaudalis (PCa) and liquid chromatography (Waters Corpo-
in multiple signaling pathways can be iliocaudalis (ICa), were harvested.17 ration nano ACQUITY, Milford, MA)
studied concurrently. Proteomics pro- Tibialis anterior (TA), a hind limb using a 20 cme75 mm ID glass capillary
vides direct assessment of the biologically skeletal muscle, served as a nonpelvic packed with 2.5 mm C18 (130) CSHTM
relevant protein levels, which often control. beads (Waters Corporation). Peptides
only modestly correlate with mRNA were eluted from the C18 column into
expression due to posttranscriptional Sample preparation for mass the mass spectrometer using a linear
regulatory mechanisms.14e16 Here, we spectroscopy gradient (5e80%) of acetonitrile (ACN)
used an unbiased quantitative proteomic Muscles were snap frozen in liquid ni- at a flow rate of 250 mL/min for 1 hour.
approach to accomplish the following trogen and sectioned on Leica Cryostat The buffers used to create the ACN
objectives: (1) to elucidate putative (Buffalo Grove, IL) to mechanically gradient were as follows: buffer A (98%
mechanisms accountable for pregnancy- disrupt the tissue. Tissue was suspended H2O, 2% ACN, 0.1% formic acid, and
induced adaptations of PFMs, and (2) in 1% sodium dodecyl sulfate in 0.005% trifuoroacetic acid), and buffer B

1.e2 American Journal of Obstetrics & Gynecology MONTH 2019

ajog.org
(100% ACN, 0.1% formic acid, and
0.005% trifuoroacetic acid). MS/MS
data were acquired in a data-dependent
manner in which the MS1 data were
acquired for 250 milliseconds at m/z of
400e1250 Da and the MS/MS data were
acquired from m/z of 50e2000 Da. The
independent data acquisition parameters
were as follows: MS1-TOF acquisition
time of 250 milliseconds, followed by 50
MS2 events of 48 milliseconds acquisi-
tion time for each event. The threshold
to trigger MS2 event was set to 150
counts when the ion had the charge
state þ2, þ3, and þ4. The ion exclusion
time was set to 4 seconds. Peptide iden-
tification and label-free quantification
analysis were carried out using Peaks
Studio 8.5 software (Bioinformatics So-
lutions Inc, Waterloo, ON, Canada).
Statistical and functional analysis
For power analysis, the effect size and
standard deviation are needed a priori,
which is not possible for exploratory
studies such as this one, as no prior data
are available. We opted for a resource
equation approach with repeated-
measures analysis of variance, based on
which our sample size calculation yiel-
ded 3 animals per group. Proteomics
data were visualized using the Qlucore
Omics Explorer (Qlucore, New York,
NY). Differentially expressed peptides
were identified using the multigroup
differential expression function (equiv-
alent to analysis of variance), with a q-
value cutoff set at <0.05 and log2FC set
at 1.5. Proteomic signatures of the
pelvic floor muscles were compared to
nonpelvic limb muscle, as well as in
nonpregnant and pregnant states.
Significantly different peptides were
imported into the online Database for
Annotation, Visualization and Inte-
grated Discovery (DAVID) v6.8 to iden-
tify biological pathways. P-values
and Benjamini scores (globally corrected
P value to control for familywise false
discovery rate) were generated by
DAVID using previously established
methods.19e23
Results
We assessed differential peptide expres-
sion in PFMs (PCa, ICa, C) and TA,
GYNECOLOGY Original Research
FIGURE 1
Principal component analysis of tryptic peptides from the rat pelvic floor
muscles and tibialis anterior
Unsupervised clustering of the data demonstrating a clear separation between samples from
nonpregnant (NP) (spheres) and pregnant (P) (squares) rats along principal component 1 (PC1).
Clustering separated pelvic floor muscles (pubocaudalis (PCa), blue; iliocaudalis (ICa), green; and
coccygeus (C), orange) from tibialis anterior (TA) (black) along principal component 2 (PC2).
Burnett et al. Proteomic alterations during pregnancy in rat pelvic muscles. Am J Obstet Gynecol 2019.
procured from nonpregnant and preg- specific expression pattern for each pro-
nant rats using principal component tein. Using this approach, we identified 2
analysis. Principal component 1, which categories of proteins, as demonstrated in
accounted for 12% of the variability Figure 3B: (1) proteins with high expres-
among samples, drew a clear separation sion in nonpregnant rats that undergo
between samples from nonpregnant and minimal changes during pregnancy; and
pregnant animals (Figure 1). Principal (2) proteins with low expression in
component 2 segregated PFMs from TA, nonpregnant animals that undergo large
accounting for an additional 10% vari- changes during pregnancy.
ability between samples (Figure 1). A total We focused on the proteins in the
of 63 peptides had significantly different second category, as they had the great-
expression between nonpregnant and est changes in expression during preg-
pregnant states. Of these, 9 proteins were nancy. Of the 5 proteins identified,
increased and 8 were decreased specif- none have been previously well char-
ically in PFMs of the pregnant rats, acterized in skeletal muscle. However,
whereas 17 were increased and 13 were two proteins (S100a6 and Camk2b)
decreased in the pregnant group in all have known functions in cell prolifera-
muscles examined. This allowed the tion, which was intriguing in the
identification of 4 separate clusters as context of sarcomerogenesis that occurs
follows: (1) proteins increased only in during pregnancy selectively in
PFMs in pregnancy; (2) proteins PFMs.12,13 S100a6 is a member of the
increased in both PFMs and TA, in S100 family of proteins, known to
pregnancy; (3) proteins decreased in both regulate cell proliferation and differen-
PFMs and TA in pregnancy; and (4) tiation in a calcium dependent
proteins decreased only in PFMs in fashion.24 Camk2b has been shown to
pregnancy (Figure 2). prevent apoptosis in hippocampal
neurons.25 Although these proteins
Cluster 1 seem unrelated, their concurrent
Pathway analysis of the proteins change during pregnancy likely con-
comprising cluster 1 (Figure 3A), to our tributes to the sarcomerogenesis of
surprise, did not identify any pathways PFMs. S100a6, which is known to
containing more than 1 of these proteins. stimulate cell proliferation, may stimu-
Consequently, we decided to examine the late resident muscle stem cells, termed
MONTH 2019 American Journal of Obstetrics & Gynecology 1.e3
Original Research GYNECOLOGY ajog.org

Specific expression patterns for each

FIGURE 2
protein were examined and demon-
Heat map of tryptic digest peptides from rat pelvic floor muscles (PFMs) and
strated 2 patterns. Metabolic enzymes
tibialis anterior (TA)
and Complex I components were
expressed in high levels in nonpregnant
animals with a small increase in preg-
nancy. On the other hand, components
of Complex V had lower expression in
nonpregnant animals and greater in-
crease in expression during pregnancy,
as illustrated in Figure 4C. Overall, we
identified 9 proteins critical for cellular
metabolism and regulation of mito-
chondrial adenosine triphosphate (ATP)
generation that were increased during
pregnancy, suggesting an upsurge in
muscle metabolic activity overall, and
mitochondrial respiration in particular.

Cluster 3
A total of 13 proteins were significantly
decreased during pregnancy in all mus-
cles examined (Figure 5A). Pathway
analysis identified 2 pathways, which
included only 4 of the proteins in this
cluster (Figure 5B). Thus, we examined
protein-specific expression changes to
identify proteins with the largest de-
creases in expression during pregnancy.
All members of this cluster had similar
Heat map demonstrating 4 major gene clusters (black boxes). Cluster 1: proteins increased in baseline expression in nonpregnant an-
pregnancy only in PFMs. Cluster 2: proteins increased in in pregnancy in both PFMs and TA. Cluster imals; however 8 proteins had a dramatic
3: proteins decreased in pregnancy in both PFMs and TA. Cluster 4: proteins decreased in pregnancy decrease in expression in the pregnant
only in PFMs. Colors represent relative quantification of protein abundance in log2 intensity scale, group. Many of these proteins are
with blue indicating the lowest and red the highest protein expression. Gene names encoding involved in cell growth (Rpf2, Smc2,
differentially expressed proteins are listed on the right side of the heatmap. Tb1d1) and cell signaling (Celsr3, Scyl2),
C, coccygeus; ICa, iliocaudalis; PCa, pubocaudalis.
and 1 protein, Camk2g, has a known
Burnett et al. Proteomic alterations during pregnancy in rat pelvic muscles. Am J Obstet Gynecol 2019.
function in skeletal muscle: regulation of
calcium release from the sarcoplasmic
reticulum.26 Overall, we found that
multiple cell growth and signaling pro-
satellite cells, that are imperative for unrelated to skeletal muscles, such as teins were decreased, suggesting restric-
sarcomerogenesis, whereas Camk2b Parkinson, Huntington, and Alzheimer tion of cell growth and muscle fiber size
may limit apoptosis in the same cellular disease pathways, identified compo- during pregnancy.
compartment. nents of the mitochondrial electron
transport chain, including Atp5f1b Cluster 4
Cluster 2 (Complex I), and Ndufs1 and Ndufa5 A total of 7 proteins were significantly
A total of 17 proteins were significantly (Complex V). These proteins were also decreased in pregnancy only in PFMs
increased in all muscles examined dur- identified as part of the oxidative (Figure 6A). Pathway analysis identified
ing pregnancy (Figure 4A). Using DA- phosphorylation pathway and meta- 5 pathways that contained 5 of these
VID analysis, we identified 8 pathways bolic pathway. proteins (Figure 6B). Interestingly,
that included 9 of these proteins The metabolic pathway also included protein-specific expression patterns
(Figure 4B). Closer examination of enzymes critical for cellular respiration showed that all proteins exhibited a
these pathways revealed enrichment in and energy production, including com- similar decrease in expression, as illus-
metabolic regulation and mitochondrial ponents of the malate shuttle (Got2, trated in Figure 6C. Thus, we focused on
function. Many pathways seemingly Mdh2) and glycolysis (Gpi, Ak1). the pathways identified, as the pattern of

1.e4 American Journal of Obstetrics & Gynecology MONTH 2019

ajog.org GYNECOLOGY Original Research

FIGURE 3
Proteomic analysis of cluster 1 proteins that are increased in pregnancy only in the pelvic floor muscles

A, Significantly increased proteins in the pelvic floor muscles during pregnancy with difference in relative quantification of protein expression derived from
peptide intensity in coccygeus (C), iliocaudalis (ICa), pubocaudalis (PCa) and nonpelvic limb muscle, tibialis anterior (TA) expressed in log2. B, Graphical
representation of protein expression in the pelvic floor muscles (pink) and nonpelvic muscle (black) expressed in log2. P values derived from 2-way
analysis of variance, followed by Tukey’s range post hoc pairwise comparisons with significance level set to 5%.
Burnett et al. Proteomic alterations during pregnancy in rat pelvic muscles. Am J Obstet Gynecol 2019.

expression did not vary between the
proteins in this cluster. Of 5 pathways, 4
included components of mitochondrial
Complex III (Uqcrh, Uqcrb). Additional
proteins identified in these pathways
included members of the troponin
complex (Tnnc2 and Tnnt3). The
troponin complex, comprising 3 pro-
teins (troponin I, T, and C), is integral
for skeletal and cardiac muscle active
contraction. Muscle force production in
skeletal muscles is controlled primarily
by changes in intracellular calcium
concentration that alter binding of the
troponins. Tnnc2 and Tnnt3 correspond
to troponin C and T isoforms found in
fast skeletal muscle.27 Parvalbumin
(Pvalb) is also found almost exclusively
in fast-contracting muscles; it accelerates
the contractionerelaxation cycle of fast
twitch muscle by speeding the rate of
relaxation via calcium shuttling.28
Overall, we found that multiple fast
skeletal muscle protein isoforms, which
are involved in regulation of muscle
contraction, are decreased in PFMs in
pregnancy, suggesting a transition from
fast to slow fiber type.
Comment
This is the first study to examine the
molecular mechanisms accountable for
pregnancy-induced adaptations in rat
pelvic floor muscles. There are 3 primary
findings of our study. First, pregnancy
fundamentally alters protein expression
in all skeletal muscles examined. Second,
protein expression in PFMs is different
from that of nonpelvic skeletal muscles.

MONTH 2019 American Journal of Obstetrics & Gynecology 1.e5

Original Research GYNECOLOGY ajog.org

FIGURE 4
Proteomic analysis of cluster 2 proteins that are increased in pregnancy in the pelvic floor muscles and
tibialis anterior

A, Significantly increased proteins in the pelvic floor muscles and tibialis anterior during pregnancy with difference in relative quantification of protein
expression derived from peptide intensity in coccygeus (C), iliocaudalis (ICa), pubocaudalis (PCa) and nonpelvic limb muscle, tibialis anterior (TA)
expressed in log2. P values derived from 2-way analysis of variance, followed by Tukey’s range post hoc testing with significance level set to 5%. B,
Pathways identified by Database for Annotation, Visualization and Integrated Discovery (DAVID) analysis of protein significantly increased in PFMs and TA
during pregnancy with count of proteins included in each pathway (column 2). P values and Benjamini scores (globally corrected P value to control for
familywise false discovery rate) were generated by DAVID using previously established methods.19e23 C, Graphical representation of protein expression
in the pelvic floor muscles (pink) and nonpelvic muscle (black) expressed in log2.
Burnett et al. Proteomic alterations during pregnancy in rat pelvic muscles. Am J Obstet Gynecol 2019.

Third, individual pelvic floor muscles
have unique alterations in protein
expression under physiological condi-
tions of pregnancy.
With respect to the first point, we un-
covered that proteomic expression
changes in all skeletal muscles during
pregnancy are mainly related to muscle
metabolism. Known characteristics asso-
ciated with greater oxidative capacity of
skeletal muscle include increased mito-
chondrial density, increased oxidative
enzymes, and reduced fiber size.29,30 We
saw increased expression of proteins
critical for cellular metabolism and
regulation of mitochondrial ATP gener-
ation during pregnancy suggesting an
increase in metabolic activity, particularly
of mitochondrial respiration in response
to physiological conditions associated
with pregnancy. Consistent with this in-
crease in respiration, we observed a
decrease in multiple cell growth and
signaling proteins during pregnancy. This
decrease is suggestive of restriction of cell
growth and muscle fiber size needed to
facilitate aerobic respiration in highly
oxidative fibers.31
In addition to shared protein expres-
sion alterations in pregnancy, we also
identified differential protein expression
in PFMs compared to nonpelvic muscle.
These differences were evident in both
pregnant and nonpregnant states. Thus,
our data indicate that PFMs are inher-
ently different from the limb muscle
despite their structural similarity. These
findings suggest that PFMs may be
uniquely equipped to respond to the
stimuli of pregnancy with subsequent
specific proteomic alterations resulting in
muscle adaptations.

1.e6 American Journal of Obstetrics & Gynecology MONTH 2019

ajog.org GYNECOLOGY Original Research

FIGURE 5
Proteomic analysis of cluster 3 proteins that are decreased in pregnancy in the pelvic floor muscles and
tibialis anterior

A, Significantly decreased proteins in the pelvic floor muscles and tibialis anterior during pregnancy with difference in relative quantification of protein
expression derived from peptide intensity in coccygeus (C), iliocaudalis (ICa), pubocaudalis (PCa), and nonpelvic limb muscle, tibialis anterior (TA)
expressed in log2. P values derived from 2-way analysis of variance, followed by Tukey’s range post hoc testing with significance level set to 5%. B,
Pathways identified by Database for Annotation, Visualization and Integrated Discovery (DAVID) analysis of protein significantly decreased in PFMs and TA
during pregnancy with count of proteins included in each pathway (column 2). P values and Benjamini scores (globally corrected P value to control for
familywise false discovery rate) were generated by DAVID using previously established methods.19e23 C, Graphical representation of protein expression
in pelvic floor muscles (pink) and nonpelvic muscle (black) expressed in log2.
Burnett et al. Proteomic alterations during pregnancy in rat pelvic muscles. Am J Obstet Gynecol 2019.

Sarcomerogenesis, or addition of sar-
comeres in series, that increases resting
fiber length is the major structural
pregnancy-induced adaptation identified
in rat PFMs thus far.11,13 Sarcomero-
genesis is a highly regulated process of
sarcomere assembly and addition within
muscle fibers associated with increase in
myonuclei number.32e34 During these
processes, quiescent resident muscle stem
cells (satellite cells) become activated and
progress through the myogenic
lineage.35e37 Here we show a higher
expression of proteins known to increase
cell proliferation and to decrease
apoptosis in PFMs during pregnancy.
Such changes appear favorable for pro-
moting satellite cell expansion, in turn
facilitating sarcomerogenesis of PFMs
that occurs under the physiological con-
ditions of pregnancy.
We also discovered a decrease in mul-
tiple fast skeletal muscle protein isoforms
in PFMs in pregnancy, supporting a
transition from fast (glycolytic) to slow
(oxidative) fiber phenotype. This fiber
phenotype transition is unique to PFMs
but complementary to the overall changes
observed in all skeletal muscles examined,
as slow fiber type is associated with
increased aerobic respiration. These novel
discoveries set the stage for the future
studies aimed at comparing PFMs’
metabolic function under nonpregnant
and pregnant conditions.
Inherent limitations of our study
include the use of an animal model to
simulate human conditions. Conse-
quently it is unknown whether these
specific protein expression changes
occur in women during pregnancy.
However, because of the inability to

MONTH 2019 American Journal of Obstetrics & Gynecology 1.e7

Original Research GYNECOLOGY ajog.org

FIGURE 6
Proteomic analysis of cluster 4 proteins that are decreased in pregnancy in the pelvic floor muscles

A, Significantly decreased proteins in the pelvic floor muscles during pregnancy, with difference in relative quantification of protein expression derived
from peptide intensity in coccygeus (C), iliocaudalis (ICa), pubocaudalis (PCa) and nonpelvic limb muscle, tibialis anterior (TA) expressed in log2. P values
derived from 2-way analysis of variance, followed by Tukey’s range post hoc testing with significance level set to 5%. B, Pathways identified by Database
for Annotation, Visualization and Integrated Discovery (DAVID) analysis of protein significantly decreased in PFMs during pregnancy with count of proteins
included in each pathway (column 2). P values and Benjamini scores (globally corrected P value to control for familywise false discovery rate) were
generated by DAVID using previously established methods.19e23. C, Graphical representation of protein expression in the pelvic floor muscles (pink)
and nonpelvic muscle (black) expressed in log2.
Burnett et al. Proteomic alterations during pregnancy in rat pelvic muscles. Am J Obstet Gynecol 2019.

directly sample PFMs in pregnant
women, animal models are critical to our
understanding of the molecular biology
of PFMs. The rat in particular has been
identified as a representative model for
studies of the pelvic floor during preg-
nancy and delivery.7,8,12,38 Despite more
favorable maternal pelvic-to-fetal size
ratio in the rat compared to humans, rat
PFMs demonstrate substantial pheno-
typic and functional adaptations. The
above suggests that the human PFMs
also undergo significant adaptations.
The current study serves as a basis for
identification of possible candidate
modulators of pregnancy-induced ad-
aptations. Whether these modulators
directly alter pelvic organ prolapse risk is
currently unknown. However, our future
goals include perturbing these signaling
pathways to directly examine their pro-
tective role against the untoward effects
of birth injury.
The diversity and extent of hormonal
and physiological alterations associated
with pregnancy that alter functionality of
virtually every organ system represent
one of the most striking nonpathological
transformations observed in
nature.39e41 Our data support the dra-
matic impact of pregnancy on skeletal
muscles, as the pregnant state accounts
for the highest degree of variability in
muscle protein expression. Future
studies are needed to validate and
localize the proteomic changes observed
in this study, and to identify the
specific stimuli during pregnancy that
direct protein expression to achieve
protective adaptations of the pelvic floor
muscles. n 4. Brooks SV, Zerba E, Faulkner JA. Injury to
Acknowledgments
This work was supported by grants from the
National Institute of Health (NICHD):
R01HD092515 and R21HD094566. We
gratefully acknowledge Majid Ghassemian, PhD,
Director of the Biomolecular/Proteomics Mass
Spectrometry Facility at the University of Cali-
fornia, San Diego, for his expertise and assis-
tance with the collection and analysis of mass
spectroscopy data.
References
1. Wu JM, Vaughan CP, Goode PS, et al.
Prevalence and trends of symptomatic pelvic
floor disorders in U.S. women. Obstet Gynecol
2014;123:141–8.
2. Lien KC, Mooney B, DeLancey JO, Ashton-
Miller JA. Levator ani muscle stretch induced by
simulated vaginal birth. Obstet Gynecol
2004;103:31–40.
3. Hoyte L, Damaser MS, Warfield SK, et al.
Quantity and distribution of levator ani stretch
during simulated vaginal childbirth. Am J Obstet
Gynecol 2008;199:198.
muscle fibres after single stretches of passive
and maximally stimulated muscles in mice.
J Physiol 1995;488:459–69.
5. DeLancey JO, Kearney R, Chou Q,
Speights S, Binno S. The appearance of levator
ani muscle abnormalities in magnetic resonance

1.e8 American Journal of Obstetrics & Gynecology MONTH 2019

ajog.org
images after vaginal delivery. Obstet Gynecol
2003;101:46–53.
6. Alperin M, Feola A, Duerr R, Moalli P,
Abramowitch S. Pregnancy- and delivery-
induced biomechanical changes in rat vagina
persist postpartum. Int Urogynecol J 2010;21:
1169–74.
7. Lowder JL, Debes KM, Moon DK, Howden N,
Abramowitch SD, Moalli PA. Biomechanical
adaptations of the rat vagina and supportive
tissues in pregnancy to accommodate delivery.
Obstet Gynecol 2007;109:136–43.
8. Daucher JA, Clark KA, Stolz DB, Meyn LA,
Moalli PA. Adaptations of the rat vagina in
pregnancy to accommodate delivery. Obstet
Gynecol 2007;109:128–35.
9. Alperin M, Tuttle LJ, Conner BR, et al. Com-
parison of pelvic muscle architecture between
humans and commonly used laboratory spe-
cies. Int Urogynecol J 2014;25:1507–15.
10. Stewart AM, Cook MS, Esparza MC,
Slayden OD, Alperin M. Architectural assess-
ment of rhesus macaque pelvic floor muscles:
comparison for use as a human model. Int
Urogynecol J 2017;28:1527–35.
11. Catanzarite T, Bremner S, Barlow CL, Bou-
Malham L, O’Connor S, Alperin M. Pelvic mus-
cles’ mechanical response to strains in the
absence and presence of pregnancy-induced
adaptations in a rat model. Am J Obstet Gyne-
col 2018;218:512.
12. Alperin M, Kaddis T, Pichika R, Esparza MC,
Lieber RL. Pregnancy-induced adaptations in
intramuscular extracellular matrix of rat pelvic floor
muscles. Am J Obstet Gynecol 2016;215:210.
13. Alperin M, Lawley DM, Esparza MC,
Lieber RL. Pregnancy-induced adaptations in
the intrinsic structure of rat pelvic floor muscles.
Am J Obstet Gynecol 2015;213:191.
14. Wu L, Candille SI, Choi Y, et al. Variation and
genetic control of protein abundance in humans.
Nature 2013;499:79–82.
15. Schwanhausser B, Busse D, Li N, et al.
Global quantification of mammalian gene
expression control. Nature 2011;473:337–42.
16. de Sousa Abreu R, Penalva LO,
Marcotte EM, Vogel C. Global signatures of
protein and mRNA expression levels. Mol Bio-
syst 2009;5:1512–26.
17. Bremer RE, Barber MD, Coates KW,
Dolber PC, Thor KB. Innervation of the levator
ani and coccygeus muscles of the female rat.
Anat Rec A Discov Mol Cell Evol Biol 2003;275:
1031–41.
18. Wisniewski JR, Zougman A, Nagaraj N,
Mann M. Universal sample preparation method
for proteome analysis. Nat Methods 2009;6:
359–62.
19. Dennis G Jr, Sherman BT, Hosack DA, et al.
DAVID: Database for Annotation, Visualization,
GYNECOLOGY
and Integrated Discovery. Genome Biol 2003;4:
P3.
20. Huang DW, Sherman BT, Tan Q, et al. The
DAVID Gene Functional Classification Tool: a
novel biological module-centric algorithm to
functionally analyze large gene lists. Genome
Biol 2007;8:R183.
21. Huang DW, Sherman BT, Tan Q, et al. DA-
VID Bioinformatics Resources: expanded
annotation database and novel algorithms to
better extract biology from large gene lists.
Nucleic Acids Res 2007;35(Web Server issue):
W169-75.
22. Huang da W, Sherman BT, Lempicki RA.
Systematic and integrative analysis of large gene
lists using DAVID bioinformatics resources. Nat
Protoc 2009;4:44–57.
23. Huang da W, Sherman BT, Lempicki RA.
Bioinformatics enrichment tools: paths toward
the comprehensive functional analysis of large
gene lists. Nucleic Acids Res 2009;37:1–13.
24. Bresnick AR, Weber DJ, Zimmer DB. S100
proteins in cancer. Nat Rev Cancer 2015;15:
96–109.
25. Fang M, Feng C, Zhao YX, Liu XY. Camk2b
protects neurons from homocysteine-induced
apoptosis with the involvement of HIF-1alpha
signal pathway. Int J Clin Exp Med 2014;7:
1659–68.
26. Saddouk FZ, Ginnan R, Singer HA. Ca(2þ)/
calmodulin-dependent protein kinase II in
vascular smooth muscle. Adv Pharmacol
2017;78:171–202.
27. Li MX, Hwang PM. Structure and function of
cardiac troponin C (TNNC1): implications for
heart failure, cardiomyopathies, and troponin
modulating drugs. Gene 2015;571:153–66.
28. Arif SH. A Ca(2þ)-binding protein with
numerous roles and uses: parvalbumin in mo-
lecular biology and physiology. Bioessays 2009
Apr;31(4):410–21.
29. Martin TP, Bodine-Fowler S, Roy RR,
Eldred E, Edgerton VR. Metabolic and fiber size
properties of cat tibialis anterior motor units. Am
J Physiol 1988l;255:C43–50.
30. Martin TP, Vailas AC, Durivage JB,
Edgerton VR, Castleman KR. Quantitative his-
tochemical determination of muscle enzymes:
biochemical verification. J Histochem Cytochem
1985;33:1053–9.
31. van Wessel T, de Haan A, van der
Laarse WJ, Jaspers RT. The muscle fiber type-
fiber size paradox: hypertrophy or oxidative
metabolism? Eur J Appl Physiol 2010;110:
665–94.
32. Montgomery RD. Growth of human striated
muscle. Nature 1962;195:194–5.
33. Williams PE, Goldspink G. Longitudinal
growth of striated muscle fibres. J Cell Sci
1971;9:751-67.
MONTH 2019 American Journal of Obstetrics & Gynecology
Original Research
34. Dayanidhi S, Lieber RL. Skeletal muscle
satellite cells: mediators of muscle growth during
development and implications for develop-
mental disorders. Muscle Nerve 2014;50:
723–32.
35. Suzuki T, Takaishi H, Sakata T, et al.
In vitro measurement of post-natal changes
in proliferating satellite cell frequency during
rat muscle growth. Anim Sci J 2010;81:
245–51.
36. Tatsumi R, Sheehan SM, Iwasaki H,
Hattori A, Allen RE. Mechanical stretch in-
duces activation of skeletal muscle satellite
cells in vitro. Exp Cell Res 2001;267:
107–14.
37. Tatsumi R, Liu X, Pulido A, et al. Satellite
cell activation in stretched skeletal muscle and
the role of nitric oxide and hepatocyte growth
factor. Am J Physiol Cell Physiol 2006;290:
C1487–94.
38. Alperin M, Feola A, Meyn L, Duerr R,
Abramowitch S, Moalli P. Collagen scaffold: a
treatment for simulated maternal birth injury in
the rat model. Am J Obstet Gynecol 2010;202:
589.
39. Chang J, Streitman D. Physiologic adapta-
tions to pregnancy. Neurol Clin 2012;30:781–9.
40. Hill CC, Pickinpaugh J. Physiologic changes
in pregnancy. Surg Clin North Am 2008;88:
391–401.
41. Yeomans ER, Gilstrap LC 3rd. Physiologic
changes in pregnancy and their impact on crit-
ical care. Crit Care Med 2005;33(10 Suppl):
S256–8.
Author and article information
From the Department of Obstetrics, Gynecology, and
Reproductive Sciences (Drs Burnett, Sesillo Boscolo, and
Laurent, Ms Wong, and Dr Alperin), University of Cali-
fornia, San Diego, San Diego, CA; Division of Maternal-
Fetal Medicine (Dr Laurent), Department of Obstetrics,
Gynecology, and Reproductive Sciences, University of
San Diego, San Diego, CA; Division of Female Pelvic
Medicine and Reconstructive Surgery (Dr Alperin),
Department of Obstetrics, Gynecology, and Reproductive
Sciences, University of California, San Diego, San Diego,
CA.
Received Dec. 21, 2018; revised March 16, 2019;
accepted April 24, 2019.
This study was conducted in San Diego, CA.
The authors report no conflict of interest.
This work was supported by grants from the National
Institute of Health (NICHD): R01HD092515 and
R21HD094566.
Portions of this work have been presented at American
Urogynecologic Society’s 39th Annual Scientific Meeting,
PFD Week, Chicago, IL, October 9e13, 2018.
Corresponding author: Marianna Alperin, MD, MS.
malperin@ucsd.edu
1.e9